Research On Scheduling Problem For Group-Process Hybrid Flow Shop

As the development of market competition severely in recent years, customer demands are transformed to diversification. Some traditional flow shop production modes are changed gradually from single brand, large batch, and simple production process to multiple brands, small batch, and complex production process. According to the needs of enterprise and customer, group-process need to be implemented depending on the characteristics of raw materials, therefore traditional flow shop production mode need to be transformed into group-process hybrid flow shop. The change of production mode makes the traditional workshop scheduling method no longer applicable. This thesis takes the group-process hybrid flow shop problem of a large cigarette factory as the research background, and then this kind of problems is researched deeply.The domestic and international relevant research in accordance with simulation optimization, flow shop scheduling with production auxiliary time, hybrid flow shop scheduling, multi-objective workshop scheduling problem and group-process hybrid flow shop scheduling problem completely summarized, and points out the problems and object in the current research.A mathematical model is established in order to research with group-process hybrid flow shop scheduling problem. The characteristics of this problem can be summarized: multi-time factor, intermediate storage/blending, complicated process, multi-object. The group-process hybrid flow shop scheduling optimization framework is set up to solve this problem depend on these characteristics. This framework combines the object-oriented discrete event modeling and heuristic evolutionary algorithm, and goes through the full paper as the workshop scheduling problem solving methods.According to multiple time factors workshop scheduling problem, the research considers delivery time among machines, and the setup time in different machine of flow shop scheduling problem. Simulation optimization model which is built to solve the problem is taken the simulation optimization theory framework as instruction. Real workshop production process is reflected through the object-oriented simulation modeling. A new population initialization method which is improved the traditional genetic algorithm genetic operator is presented. The research conclusion is concluded from the example analysis.For intermediate storage/blending characteristics workshop scheduling problem, the problem with intermediate storage/blending strategy of hybrid flow assembling line is studied. The objective function is non-process storage time and early/late time. The linear weighted sum method is used to deal with two optimization objectives. The material sequence and machine assignment is solved from combining the natural number coding and classical dispatching rules, and improves the evolution genetic operator, at the same time actual case is analyzed.In the face of this complicated process workshop scheduling problem, a flexible tobacco strip manufacturing system model is established based on the discrete event modeling method, and analyzed the problems of this production process, system composition and key equipment. Meanwhile the system time simulation method and operation process of the production plan to the finished product are presented. A certain enterprise flexible tobacco strips manufacturing system model is established from using this method. The simulation model is analyzed based on the actual flexible tobacco strips manufacturing process.Based on the above research, the flexible tobacco strips scheduling system is built depending on a large cigarette production enterprise’s group-process. This thesis proposes the multi-objective optimization function and constraint conditions of this problem, and also proposes an optimization model with a Pareto ranking theory. The new initial population strategy is presented to improve the quality of initial population. Improving the genetic operators in evolution process, using local search strategy and a new elitist strategy, these improvements ensure that the quality of Pareto optimal solution in the process of evolution. Finally, the examples are analyzed based on tobacco group-process design module and the actual production module.